With the increased expansion of network scale and demands for high-quality services, at present, the optical network has higher and higher demands for network survivability. The network survivability refers to that the network can still preserve acceptable service quality grade in the case of failures. The APS operations and the recovery operations are main means for improving optical network survivability and supporting the demands for QoS (Quality of Service) of service transfer.
The basic idea of the APS operations is to pre-configure a protection channel for services. In the case of fault occurring in the working channel, the working channel detection unit of corresponding node detects an alarm and reports it to the protection protocol unit of the present node; the protection protocol unit starts up an APS operation after receiving the working channel alarm, that is, the protection protocol unit runs the configured protection protocol algorithm and also exchanges signals with protection protocol units of other nodes in the protection channel; then the protection protocol unit of each node in the protection channel sends an switching instruction to the execution units of the present node respectively; finally, the execution units of each node in the protection channel execute the APS operation and switch the service over to the protection channel to ensure the normal operation of services. APS operations bring about the advantages of rapid switching and short service interruption time, but the network resources utilization rate is relatively low.
The recovery operations do not pre-configure protection channel for services. In the case of fault occurring in the working channel, the working channel detection unit of corresponding node detects an alarm and reports it to the recovery protocol unit of the present node; the recovery protocol unit starts up the recovery operation after receiving the working channel alarm, that is, this recovery protocol unit recalculates a new channel for services in the idle resources of current network, and also exchanges signals with recovery protocol units of other nodes in the new channel; then the recovery protocol unit of each node in the new channel sends a switching instruction to the execution units of the present node respectively; finally, the execution units of each node in the new channel execute recovery operation and switch services over to the new channel to ensure the normal operation of services. The recovery operation has a relatively high utilization rate of network resources, but it requires real-time computation of service channel and thus the service fault time is relatively long.
As for the service requiring higher QoS, the service is often provided with APS functions and recovery functions. As illustrated in FIG. 1, the device comprises a working channel detection unit, a protection protocol unit and a recovery protocol unit, wherein when the working channel detection unit detects a working channel alarm, it reports the alarm to the protection protocol unit and the recovery protocol unit; the protection protocol unit starts up an APS operation immediately after receiving the working channel alarm, while the recovery protocol unit waits for a pre-set time period (holdoff) after receiving the working channel alarm and then checks whether the working channel detection unit still detects an alarm or not, wherein if yes, it starts up the recovery operation, otherwise the recovery operation needs not to be started up because the service operation is normal. Due to lack of signaling interaction between the protection protocol unit and the recovery protocol unit, in the case of the APS function failure, e.g. when the working channel and protection channel of the service both fail and the recovery operation shall be started up immediately, the recovery protocol unit can not start up the recovery operation until waiting for the time of period (holdoff), such that the service fault time is increased by the time of holdoff, which increases service interruption time.